Selected Papers from: Communications & Information Technology 2008, Circuits, Systems and Signals 2008,
Applied Mathematics, Simulation, Modelling 2008, Marathon Beach, Attica, Greece, June 1-3, 2008
The influence of the sheet thickness on
springback effect in case of TWBs forming
Aurelian Albut
AbstractThis paper presents the results obtained by numerical
simulation regarding springback phenomenon of a part manufactured
from tailor welded blanks. The final shape of the formed part is
seriously affected by springback phenomenon. This paper work is
trying to prove the important role the metal sheet thickness on the
springback effect. The part has different springback values for each
material from the welded assembly structure. The influence of the
sheet thickness on the tailor welded stripes springback is examined
by finite element method using Abaqus Standard for forming process
and Abaqus Explicit for springback of the obtained part.
Keywordstailor welded blanks, metal sheet thickness, springback.
I. INTRODUCTION
ecently, the automobile industries have been trying to
develop various types of model and high-quality low-cost
cars to meet the customers requirements and to find new
ways of establishing this goal effectively. For the purpose of
achieving the above presented objectives, different methods
using various welding processes (such as laser-welding, mash
seam-welding processes, etc.) were developed. A tailor
welded blank consists of two or more sheets that have been
welded together in a single plane prior to forming. The sheets
can be identical, or can have different thickness, mechanical
properties or surface coatings [1, 2, 3, 4]. Thickness of tailor
welded sheets plays an important role in sheet metal forming
since fracture, wrinkling and weak spots are strongly
influenced by material behaviour.
The techniques of numerical analysis applicable for sheet
metal forming have been considerably developed for the last
several years. However, accurate prediction of the springback
remains elusive [5, 6]. In finite element method (FEM) models
of metal forming, the roughness has usually been assumed to
be constant; even though it is commonly observed that sheet
drawn under tension over a tool radius results in the surface
becoming shiny, indicating a major change in surface
morphology.
Many studies present a wide range of information about the
formability and failure patterns of welded blanks [7]. A wide
range of information about the formability and failure patterns
of tailor-welded blanks and the springback of non-welded
Manuscript received May 15, 2008. This work was supported in part by
the Romanian Ministry of Education and Research.
Aurelian Albut is with Faculty of Engineering, University of Bacau,
Bacau600115 ROMANIA (phone: +[Link]; fax: +[Link];
e-mail: albut_aurelian@ [Link]).
ISBN: 978-960-6766-69-5
147
sheet metal parts has been presented. However, the springback
characteristics of tailor-welded blanks have hardly been found
[8, 9]. Published results on springback prediction of tailor
welded blanks are minimal [11].
Since the springback is also affected by the material
properties, such as Youngs modulus and initial yield stress,
the process design for tailor-welded blanks is more
complicated than for a homogenous metal sheet. Though novel
approaches relating to the formality of tailor-welded blanks are
available, the change of springback due to the characteristic of
each process should be verified by finite element method [11].
In this study, the tailor welded blanks with two types of
material having the same thickness, are used to investigate
springback characteristics in U-shape forming.
Springback
angle
Bending
angle
Bended part
angle
Fig. 1 scheme of the bending process and springback
Springback (Fig. 1) is mainly influenced by the sheet
thickness, the punch and die profile radii, initial clearance
between punch and die, friction conditions, rolling direction of
the materials, blankholder force, material properties (elastic
modulus, Poissons coefficient, constitutive behaviour in
plastic field) etc. The purpose of this study was to investigate
the sheet thickness influence on the springback effect of the
tailor-welded stripes. To achieve this goal, simulation tests
were carried with different sheet thickness of the laser welded
assembly.
II. MECHANICAL PROPERTIES OF MATERIALS
To obtain the mechanical properties of the base materials
and of the welding line, tensile tests were performed on a
universal mechanical testing machine, equipped with Hottinger
force cells of 5 tf and a Hottinger Baldwin electronic
measurement system for PCs type Spider 8. The data
acquisition, processing and visualisation were performed using
Catman Express software. The measurement of specific strains
for determination of stress - strain curves was performed using
an Epsilon extensometer for a strain rate of 0.1 s-1.
ISSN: 1790-2769
Selected Papers from: Communications & Information Technology 2008, Circuits, Systems and Signals 2008,
Applied Mathematics, Simulation, Modelling 2008, Marathon Beach, Attica, Greece, June 1-3, 2008
To determine the mechanical properties of the welding line,
from the original TWB, a 4 mm wide stripe which includes the
welding line, has been removed using EDM wire cutting (Fig. 2).
The materials microstructure has been analysed using a
metallographic microscope with a magnification of 100X for
base materials and 500X for welding line. The materials have
uniform, typical microstructure with fine grain. The
microstructure is shown in Fig. 4.
Fig. 2 sample used to determine the mechanical properties of
the welding line
The flow stress and true total strain were calculated for each
recorded couple of the force and displacement. The total strain
was decomposed on elastic strain and plastic strain using
determined Young module.
To obtain a better accuracy of FEM modeling, especially in
the range of small deformation, there was resigned the
functional stress strain curves in favor of the stress strain
curves in the numerical form.
The mechanical properties of FEPO steel and E220 steel
determined for 0 and 90 are presented in table no 1. In the
table below are presented also, the mechanical properties of
the welding line.
TABLE 1
MECHANICAL PROPERTIES
Property
Yield strength
Rp0,2 [MPa]
Tensile strength
Rm [MPa]
Percentage elongation
after fracture A80 [%]
FeP0
E220
203
268
90
209
271
381
458
90
382
456
35,3
90
31,1
weld
RD
RD
FEPO
20,4
90
31,0
19,8
Strain-hardening
coefficient n
0,222
0,190
90
0,225
0,180
Plastic strain ratio
r
1,860
1,420
90
2,430
1,650
Poisson's ratio
0,286
0,297
90
0,294
0.232
200825
204271
90
206467
203772
17,3
RP
RP
Fig. 5 springback parameters
The initial dimensions of the sheet were 350 mm length and
30 mm width, with thickness variation from 0,8 till 1,3 mm.
For determination of the sheet thickness influence on
springback phenomenon, the simulations have been done
under the following conditions: the sample model is loaded
with the material characteristics corresponding to 0o rolling
direction; blank holder force F = 15 kN; the friction coefficient
between the specimen and tools surfaces is 0.1, the thickness
of the sample model are 0.8, 0.9, 1.0, 1.1, 1.2 and 1.3 mm.
Figure 6 presents the initial and the final state of the forming
process in case of the U shaped part made from TWBs.
1,29
0,278
203253
500
450
400
E220
300
FEPO
250
weld
200
150
100
50
0
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
Strain (mm/mm)
Fig. 6 geometrical model
Fig. 3 stress strain curves for 0
ISBN: 978-960-6766-69-5
E220
31,8
Stress (MPa)
A. Simulation methodology
The simulation of the U - shape part draw bending was
made by using the ABAQUS software. The analyzed
geometrical parameters are sidewall radius and springback
angles 1 and 2 (Fig. 5). The simulations considered a plane
strain state. The material was modelled as elastic-plastic, the
plastic behaviour being modelled as anisotropic using the
Hills quadratic anisotropic yield criterion.
2
417
350
III. ANALYSIS BY SIMULATION
252
Elongation for max.
Load Agt [%]
Young modulus
E [MPa]
E220
FEPO
weld
Fig. 4 the material microstructure
148
ISSN: 1790-2769
Selected Papers from: Communications & Information Technology 2008, Circuits, Systems and Signals 2008,
Applied Mathematics, Simulation, Modelling 2008, Marathon Beach, Attica, Greece, June 1-3, 2008
The sheet was modelled as deformable body with 400 shell
elements (S4R) on one row with 5 integration points through
the thickness. The tools (punch, die and blankholder) were
modelled as analytical rigid body. This modelling will lead to
a reduced calculus efforts and a better contact behaviour. The
rigid body movements were controlled by the reference points.
The boundary conditions imposed to the tools were intended
to describe as accurate as possible the experimental conditions.
A modified Coulomb friction law combined with penalty
method was used to describe the contact condition.
Parts having different thickness are affected in a dissimilar mode
by the springback phenomena. The values of springback
parameters are recorded in table 2.
TABLE 2
SPRINGBACK PARAMETERS
FEPO
Metal
Sidewall radius
sheet Angle 1 [grd] Angle 2 [grd]
[mm]
thickness TheoreticMeasuredTheoreticMeasuredTheoreticMeasured
value value value value value value
[mm]
0.8
100.6
14.9
205.0
0.9
99.7
14.1
307.6
1.0
99.0
13.4
385.8
90
0
1.1
96.8
10.3
1524.0
1.2
95.1
7.6
2875.5
1.3
93.6
5.5
3347.3
B. Simulation results
The values of springback parameters resulted from the
simulation tests are graphical represented in figures 7, 8 and 9.
104
E220
Springback angle 1
102
Metal
Sidewall radius
sheet Angle 1 [grd] Angle 2 [grd]
[mm]
thickness TheoreticMeasuredTheoreticMeasuredTheoreticMeasured
value value value value value value
[mm]
0.8
102.8
19.3
188.4
0.9
101.0
17.5
265.7
1.0
100.1
15.6
322.9
90
0
1.1
98.2
11.8
818.5
1.2
96.7
9.5
2106.6
1.3
95.6
8.3
3071.5
100
E220
98
96
FEPO
94
92
0.7
0.8
0.9
1.1
1.2
1.3
1.4
Metal sheet thickness
IV. CONCLUSION
Fig.7 variation of 1 angle
From the analysis of figure 10 and table 2, the following
observations can be presented concerning the influence of the
sheet thickness on springback parameters:
the modification of sheet thickness leads to important
variations of springback parameters;
increasing of the sheet thickness results in reduction of the
springback effect, the final geometry of the formed part is
closer to the ideal part shape;
variation of the springback phenomenon proportional with
sheet thickness is observed for both areas of the part;
for sheet thickness over 1.0 mm the sidewall radius
increases so much, that it can be considered a straight line;
the springback parameters from the part zone made from
FEPO steel record smaller values for both angles 1 and 2
with respect to the E220 part area;
the part area made by E220 present a springback intensity
higher that the part area made from FEPO steel.
25
Springback angle 2
20
15
E220
10
FEPO
5
0
0.7
0.8
0.9
1
1.1
1.2
Metal sheet thickness
1.3
1.4
Fig. 8 variation of 2 angle
4000
Sidewall curvature radius
3500
REFERENCES
3000
[1]
2500
FEPO
2000
E220
1500
[2]
1000
[3]
500
0
0.7
0.8
0.9
1.1
1.2
1.3
1.4
Metal sheet thickness
[4]
Fig. 9 variation of sidewall curvature radius
ISBN: 978-960-6766-69-5
149
G.Y. Li, M.J. Tan and K.M. Liew, Springback analysis for sheet
forming processes by explicit finite element method in conjunction with
the orthogonal regression analysis, Int. Journ. of Solids and Structures,
36, 1999
L. Papeleux and J.P. Ponthot, Finite element simulation of springback
in sheet metal forming, J. of Mat. Proc. Tech., 125-126, 2002
N. Yamamura, a.o., Springback simulation by the static explicit FEM
code, using a new algorithm for canceling the non-equilibrated forces,
Simulations of Materials Processing: Theory, Methods and
Applications, Swets & Zeitlinger, Lisse, 2001
F. Pourboghrat and E. Chu, Prediction of spring-back and side-wall
curl in 2-D draw bending, J. of Mat. Proc. Tech., 50, 1995
ISSN: 1790-2769
Selected Papers from: Communications & Information Technology 2008, Circuits, Systems and Signals 2008,
Applied Mathematics, Simulation, Modelling 2008, Marathon Beach, Attica, Greece, June 1-3, 2008
1.3mm
FEPO
1.2mm
E220
1.1mm
1.0mm
0.9mm
0.8mm
Fig. 10 final shape of the TWBs affected by springback phenomenon
[5]
F. Pourboghrat and E. Chu, Springback in plane strain stretch/draw
sheet forming, Int. J. Mech. Sci., vol. 36, no. 3, 1995
[6] J. K. Lee, B. K. Chun and H. Y. Kim, Numerical investigation of tailor
welded blanks forming and springback, Simulation of Material
Processing, p 729-734, 2001
[7] G. Liu, Z. Lin, W. Xu and Y. Bao, Variable blankholder force in ushaped part forming for eliminating springback error, J. of Mat. Proc.
Tech., vol. 120, 2002
[8] M. Samuel, Experimental and numerical prediction of springback and
side wall in u-bending of anisotropic sheet metals, J. of Mat. Proc.
Tech. 382-393, 2000.
[9] B. Chirita, G. Brabie, Experimental analysis of different influences on
springback of parts formed by U-bending, 7th International
Research/Expert Conference Trends in the Development of Machinery
and Asociated Technology TMT 2003, Lloret del Mar, Barcelona,
Spain, 15-16 September 2003.
[10] B. Chirita, Factors of influence on the springback of formed metal
sheets, ESAFORM The 5th International Conference on Material
Forming, Krakow, Poland, 2002.
[11] S.S. Han and K.C. Park, An investigation of the factors influencing
springback by empirical and simulative techniques, Numisheet99,
Besancon, France, 13-17 September 1999.
ISBN: 978-960-6766-69-5
150
ISSN: 1790-2769